fix single_cpu tests

Author: jorisvandenbossche

pandas/tests/apply/test_numba.py

@@ -1,6 +1,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_string_dtype
+
 import pandas.util._test_decorators as td
 
 from pandas import (
@@ -17,6 +19,7 @@ def apply_axis(request):
     return request.param
 
 
+@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
 def test_numba_vs_python_noop(float_frame, apply_axis):
     func = lambda x: x
     result = float_frame.apply(func, engine="numba", axis=apply_axis)
@@ -40,6 +43,7 @@ def test_numba_vs_python_string_index():
     )
 
 
+@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
 def test_numba_vs_python_indexing():
     frame = DataFrame(
         {"a": [1, 2, 3], "b": [4, 5, 6], "c": [7.0, 8.0, 9.0]},

pandas/tests/io/json/test_pandas.py

@@ -1119,7 +1119,7 @@ def test_round_trip_exception(self, datapath):
         res = res.fillna(np.nan)
         tm.assert_frame_equal(res, df)
 
-    @pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
+    @pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)", strict=False)
     @pytest.mark.network
     @pytest.mark.single_cpu
     @pytest.mark.parametrize(
@@ -1422,7 +1422,6 @@ def test_read_inline_jsonl(self):
         expected = DataFrame([[1, 2], [1, 2]], columns=["a", "b"])
         tm.assert_frame_equal(result, expected)
 
-    @pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
     @pytest.mark.single_cpu
     @td.skip_if_not_us_locale
     def test_read_s3_jsonl(self, s3_public_bucket_with_data, s3so):
@@ -2022,7 +2021,6 @@ def test_json_multiindex(self):
         result = series.to_json(orient="index")
         assert result == expected
 
-    @pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
     @pytest.mark.single_cpu
     def test_to_s3(self, s3_public_bucket, s3so):
         # GH 28375

pandas/tests/io/pytables/test_append.py

@@ -5,6 +5,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_string_dtype
+
 from pandas._libs.tslibs import Timestamp
 from pandas.compat import PY312
 
@@ -23,7 +25,10 @@
     ensure_clean_store,
 )
 
-pytestmark = pytest.mark.single_cpu
+pytestmark = [
+    pytest.mark.single_cpu,
+    pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)", strict=False),
+]
 
 tables = pytest.importorskip("tables")
 

pandas/tests/io/pytables/test_categorical.py

@@ -1,6 +1,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_string_dtype
+
 from pandas import (
     Categorical,
     DataFrame,
@@ -14,7 +16,10 @@
     ensure_clean_store,
 )
 
-pytestmark = pytest.mark.single_cpu
+pytestmark = [
+    pytest.mark.single_cpu,
+    pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)", strict=False),
+]
 
 
 def test_categorical(setup_path):

pandas/tests/io/pytables/test_complex.py

@@ -13,6 +13,10 @@
 
 from pandas.io.pytables import read_hdf
 
+pytestmark = pytest.mark.xfail(
+    using_string_dtype(), reason="TODO(infer_string)", strict=False
+)
+
 
 def test_complex_fixed(tmp_path, setup_path):
     df = DataFrame(
@@ -61,7 +65,6 @@ def test_complex_table(tmp_path, setup_path):
     tm.assert_frame_equal(df, reread)
 
 
-@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
 def test_complex_mixed_fixed(tmp_path, setup_path):
     complex64 = np.array(
         [1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j], dtype=np.complex64
@@ -85,7 +88,6 @@ def test_complex_mixed_fixed(tmp_path, setup_path):
     tm.assert_frame_equal(df, reread)
 
 
-@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
 def test_complex_mixed_table(tmp_path, setup_path):
     complex64 = np.array(
         [1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j], dtype=np.complex64
@@ -140,7 +142,6 @@ def test_complex_across_dimensions(tmp_path, setup_path):
     tm.assert_frame_equal(df, reread)
 
 
-@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
 def test_complex_indexing_error(setup_path):
     complex128 = np.array(
         [1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j, 1.0 + 1.0j], dtype=np.complex128

pandas/tests/io/pytables/test_errors.py

@@ -5,6 +5,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_string_dtype
+
 from pandas import (
     CategoricalIndex,
     DataFrame,
@@ -22,7 +24,10 @@
     _maybe_adjust_name,
 )
 
-pytestmark = pytest.mark.single_cpu
+pytestmark = [
+    pytest.mark.single_cpu,
+    pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)", strict=False),
+]
 
 
 def test_pass_spec_to_storer(setup_path):

pandas/tests/io/pytables/test_file_handling.py

@@ -3,6 +3,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_string_dtype
+
 from pandas.compat import (
     PY311,
     is_ci_environment,
@@ -33,7 +35,10 @@
 from pandas.io import pytables
 from pandas.io.pytables import Term
 
-pytestmark = pytest.mark.single_cpu
+pytestmark = [
+    pytest.mark.single_cpu,
+    pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)", strict=False),
+]
 
 
 @pytest.mark.parametrize("mode", ["r", "r+", "a", "w"])

pandas/tests/io/pytables/test_put.py

@@ -3,6 +3,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_string_dtype
+
 from pandas._libs.tslibs import Timestamp
 
 import pandas as pd
@@ -22,7 +24,10 @@
 )
 from pandas.util import _test_decorators as td
 
-pytestmark = pytest.mark.single_cpu
+pytestmark = [
+    pytest.mark.single_cpu,
+    pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)", strict=False),
+]
 
 
 def test_format_type(tmp_path, setup_path):

pandas/tests/io/pytables/test_read.py

@@ -5,6 +5,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_string_dtype
+
 from pandas.compat import is_platform_windows
 
 import pandas as pd
@@ -24,7 +26,10 @@
 
 from pandas.io.pytables import TableIterator
 
-pytestmark = pytest.mark.single_cpu
+pytestmark = [
+    pytest.mark.single_cpu,
+    pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)", strict=False),
+]
 
 
 def test_read_missing_key_close_store(tmp_path, setup_path):

pandas/tests/io/pytables/test_round_trip.py

@@ -4,6 +4,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_string_dtype
+
 from pandas._libs.tslibs import Timestamp
 from pandas.compat import is_platform_windows
 
@@ -24,7 +26,10 @@
 )
 from pandas.util import _test_decorators as td
 
-pytestmark = pytest.mark.single_cpu
+pytestmark = [
+    pytest.mark.single_cpu,
+    pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)", strict=False),
+]
 
 
 def test_conv_read_write():

pandas/tests/io/pytables/test_select.py

@@ -1,6 +1,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_string_dtype
+
 from pandas._libs.tslibs import Timestamp
 from pandas.compat import PY312
 
@@ -25,7 +27,10 @@
 
 from pandas.io.pytables import Term
 
-pytestmark = pytest.mark.single_cpu
+pytestmark = [
+    pytest.mark.single_cpu,
+    pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)", strict=False),
+]
 
 
 def test_select_columns_in_where(setup_path):

pandas/tests/io/pytables/test_store.py

@@ -7,6 +7,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_string_dtype
+
 from pandas.compat import PY312
 
 import pandas as pd
@@ -33,7 +35,10 @@
     read_hdf,
 )
 
-pytestmark = pytest.mark.single_cpu
+pytestmark = [
+    pytest.mark.single_cpu,
+    pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)", strict=False),
+]
 
 tables = pytest.importorskip("tables")
 

pandas/tests/io/pytables/test_timezones.py

@@ -25,6 +25,10 @@
     ensure_clean_store,
 )
 
+pytestmark = pytest.mark.xfail(
+    using_string_dtype(), reason="TODO(infer_string)", strict=False
+)
+
 
 def _compare_with_tz(a, b):
     tm.assert_frame_equal(a, b)
@@ -44,7 +48,6 @@ def _compare_with_tz(a, b):
 gettz_pytz = lambda x: x
 
 
-@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
 @pytest.mark.parametrize("gettz", [gettz_dateutil, gettz_pytz])
 def test_append_with_timezones(setup_path, gettz):
     # as columns
@@ -219,7 +222,6 @@ def test_tseries_select_index_column(setup_path):
         assert rng.tz == result.dt.tz
 
 
-@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
 def test_timezones_fixed_format_frame_non_empty(setup_path):
     with ensure_clean_store(setup_path) as store:
         # index
@@ -249,7 +251,6 @@ def test_timezones_fixed_format_frame_non_empty(setup_path):
         tm.assert_frame_equal(result, df)
 
 
-@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
 def test_timezones_fixed_format_empty(setup_path, tz_aware_fixture, frame_or_series):
     # GH 20594
 
@@ -265,7 +266,6 @@ def test_timezones_fixed_format_empty(setup_path, tz_aware_fixture, frame_or_ser
         tm.assert_equal(result, obj)
 
 
-@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
 def test_timezones_fixed_format_series_nonempty(setup_path, tz_aware_fixture):
     # GH 20594
 
@@ -318,7 +318,6 @@ def test_store_timezone(setup_path):
         tm.assert_frame_equal(result, df)
 
 
-@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
 def test_dst_transitions(setup_path):
     # make sure we are not failing on transitions
     with ensure_clean_store(setup_path) as store:
@@ -339,7 +338,6 @@ def test_dst_transitions(setup_path):
             tm.assert_frame_equal(result, df)
 
 
-@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
 def test_read_with_where_tz_aware_index(tmp_path, setup_path):
     # GH 11926
     periods = 10

pandas/tests/io/test_clipboard.py

@@ -3,6 +3,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_string_dtype
+
 from pandas.errors import (
     PyperclipException,
     PyperclipWindowsException,
@@ -28,6 +30,10 @@
     init_qt_clipboard,
 )
 
+pytestmark = pytest.mark.xfail(
+    using_string_dtype(), reason="TODO(infer_string)", strict=False
+)
+
 
 def build_kwargs(sep, excel):
     kwargs = {}

pandas/tests/io/test_feather.py

@@ -5,6 +5,8 @@
 import numpy as np
 import pytest
 
+from pandas._config import using_string_dtype
+
 import pandas as pd
 import pandas._testing as tm
 from pandas.core.arrays import (
@@ -146,6 +148,7 @@ def test_path_pathlib(self):
         result = tm.round_trip_pathlib(df.to_feather, read_feather)
         tm.assert_frame_equal(df, result)
 
+    @pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
     def test_passthrough_keywords(self):
         df = pd.DataFrame(
             1.1 * np.arange(120).reshape((30, 4)),
@@ -164,6 +167,7 @@ def test_http_path(self, feather_file, httpserver):
             res = read_feather(httpserver.url)
         tm.assert_frame_equal(expected, res)
 
+    @pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)", strict=False)
     def test_read_feather_dtype_backend(self, string_storage, dtype_backend):
         # GH#50765
         df = pd.DataFrame(

pandas/tests/io/test_fsspec.py

@@ -262,6 +262,7 @@ def test_s3_protocols(s3_public_bucket_with_data, tips_file, protocol, s3so):
     )
 
 
+@pytest.mark.xfail(using_string_dtype(), reason="TODO(infer_string)")
 @pytest.mark.single_cpu
 def test_s3_parquet(s3_public_bucket, s3so, df1):
     pytest.importorskip("fastparquet")